--- title: "Prompt Analyzer" method: POST path: "/api/v1/submit/prompt-analyzer" tags: ["interactions-v1_other"] --- # Prompt Analyzer `POST /api/v1/submit/prompt-analyzer` Analyze LLM Prompt and Response ## Request body - PromptAnalyzerRequest — The request to the prompt analyzer endpoint - `prompt` string, required - `output` string - `model` string ## Response `200` Successful Prompt Analyzer response - HiddenLayerPromptAnalyzerResponse - `response` HiddenLayerPromptAnalyzerResponseMetadata - `provider` string - `model` string - `prompt` string - `output` string - `unmodified_prompt` string - `unmodified_output` string - `provider` string - `model` string - `verdict` boolean — The overall verdict of the analysis - `categories` HiddenLayerAnalysisCategories — The analysis detection categories - `unsafe_input` boolean — The input is unsafe - `unsafe_output` boolean — The output is unsafe - `prompt_injection` boolean — The input contains prompt injection - `input_dos` boolean — The input contains a denial of service attack - `input_pii` boolean — The input contains personally identifiable information - `output_pii` boolean — The output contains personally identifiable information - `input_code` boolean — The input contains code - `output_code` boolean — The output contains code - `guardrail` boolean — The input activated the upstream guardrails - `input_language` boolean — The input contains a disallowed language - `results` HiddenLayerAnalysisResults — The analysis results - `input_block_list_results` object — The input block list results - `verdict` boolean — The verdict of the input block list analysis - `matches` string[] - `elapsed_ms` number — The time in milliseconds it took to process the input block list - `prompt_injection_classifier_results` object[] - `version` number — The version of the prompt injection classifier - `verdict` boolean — The verdict of the prompt injection classifier - `probabilities` number[] - `elapsed_ms` number — The time in milliseconds it took to process the prompt injection classifier - `allow_overrides` string[] — The list of allow overrides applied to the prompt - `block_overrides` string[] — The list of block overrides applied to the prompt - `appeal` object — Appeal details for the prompt injection decision, if any - `input_dos_results` object — The input denial of service results - `verdict` boolean — The verdict of the input denial of service analysis - `elapsed_ms` number — The time in milliseconds it took to process the input denial of service - `embeddings_length` number — The length of the embeddings analyzed - `input_pii_results` object — The input personally identifiable information results - `verdict` boolean — The verdict of the input personally identifiable information analysis - `entities` string[] - `elapsed_ms` number — The time in milliseconds it took to process the input personally identifiable information - `output_pii_results` object — The output personally identifiable information results - `verdict` boolean — The verdict of the output personally identifiable information analysis - `entities` string[] - `elapsed_ms` number — The time in milliseconds it took to process the output personally identifiable information - `input_code_results` object — The input code results - `verdict` boolean — The verdict of the input code analysis - `elapsed_ms` number — The time in milliseconds it took to process the input code - `output_code_results` object — The output code results - `verdict` boolean — The verdict of the output code analysis - `elapsed_ms` number — The time in milliseconds it took to process the output code - `guardrail_results` object — The guardrail results - `verdict` boolean — The verdict of the guardrail analysis - `elapsed_ms` number — The time in milliseconds it took to process the guardrail - `refusal_classifier_results` object — The refusal classifier results - `version` number — The version of the refusal classifier - `verdict` boolean — The verdict of the refusal classifier - `probabilities` number[] - `elapsed_ms` number — The time in milliseconds it took to process the refusal classifier - `input_url_results` object — The input URL results - `urls` string[] - `elapsed_ms` number — The time in milliseconds it took to process the guardrail - `output_url_results` object — The output URL results - `urls` string[] - `elapsed_ms` number — The time in milliseconds it took to process the guardrail - `input_language_results` object — The input language results - `verdict` boolean — The verdict of the input language analysis - `language` string — Language detected in the input - `elapsed_ms` number — The time in milliseconds it took to process the input language detection - `policy` HiddenLayerAnalysisPolicy — The policy used during analysis - `block_unsafe` boolean — Block unsafe input and output - `block_unsafe_input` boolean — Block unsafe input - `block_unsafe_output` boolean — Block unsafe output - `skip_prompt_injection_detection` boolean — Skip prompt injection detection - `block_prompt_injection` boolean — Block prompt injection - `prompt_injection_scan_type` 'quick' | 'full' — The type of prompt injection scan to use - `skip_input_pii_detection` boolean — Skip input personally identifiable information detection - `skip_output_pii_detection` boolean — Skip output personally identifiable information detection - `block_input_pii` boolean — Block input personally identifiable information - `block_output_pii` boolean — Block output personally identifiable information - `redact_input_pii` boolean — Redact input personally identifiable information - `redact_output_pii` boolean — Redact output personally identifiable information - `redact_type` 'entity' | 'strict' — The type of redaction to use - `entity_type` 'strict' | 'all' — The type of entity to redact - `skip_input_code_detection` boolean — Skip input code detection - `skip_output_code_detection` boolean — Skip output code detection - `block_input_code_detection` boolean — Block input code detection - `block_output_code_detection` boolean — Block output code detection - `skip_guardrail_detection` boolean — Skip guardrail detection - `block_guardrail_detection` boolean — Block guardrail detection - `skip_input_url_detection` boolean — Skip input URL detection - `skip_output_url_detection` boolean — Skip output URL detection - `skip_input_dos_detection` boolean — Skip input denial of service detection - `block_input_dos_detection` boolean — Block input denial of service detection - `input_dos_detection_threshold` number — The threshold for input denial of service detection - `frameworks` HiddenLayerAnalysisFrameworks — The framework labels identified during analysis - `mitre` object[] - `name` string — The name of the MITRE Atlas framework label - `label` string — The label of the MITRE Atlas framework label - `owasp` object[] - `name` string — The name of the OWASP framework label - `label` string — The label of the OWASP framework label - `owasp:2025` object[] - `name` string — The name of the OWASP:2025 framework label - `label` string — The label of the OWASP:2025 framework label - `elapsed_ms` number — The time in milliseconds it took to process the request - `upstream_elapsed_ms` number — The time in milliseconds the upstream LLM took to process the request --- [API](https://skmtc.net/hiddenlayerai/apis/hiddenlayer-audit-api.md) · [All operations](https://skmtc.net/hiddenlayerai/apis/hiddenlayer-audit-api/llms.txt) · [OpenAPI document](https://skmtc-service-staging.skmtc.workers.dev/v1/apis/hiddenlayerai/hiddenlayer-audit-api/revisions/efca861e915b/schema)